LIU Yiming,HE Xiaoxi,HUANG Shixian.Research and Application of Large-scale 3D Model Accelerated Rendering Technology[J].Journal of Chengdu University of Information Technology,2021,36(03):300-304.[doi:10.16836/j.cnki.jcuit.2021.03.010]
大规模三维模型加速渲染技术的研究与应用
- Title:
- Research and Application of Large-scale 3D Model Accelerated Rendering Technology
- 文章编号:
- 2096-1618(2021)03-0300-05
- Keywords:
- large-scale scene; acceleration structure; Morton code; quadtree; BVH
- 分类号:
- TP312
- 文献标志码:
- A
- 摘要:
- 随着在游戏和仿真领域中对大规模场景的要求日益提高,为了保证一般计算机在大规模三维模型的场景中仍能够较高帧率运行,基于Morton码提出了一种场景遍历算法。该算法根据满四叉树节点编码思想对线性四叉树的Morton码进行了改进,提出了一种线性多段结构,并且结合了自底向上合并的BVH来减少遍历节点,最后利用视锥体对场景内的物体进行快速裁剪。该算法结合场景中的多层均匀网格,利用CPU的多核并行计算能力,根据上层遍历结果,快速得到下层输入,能够快速对场景中的物体进行可见性裁剪,提高计算机在大规模模型场景的渲染帧率。实验结果表明了该算法的正确性和可行性,与传统的加速结构相比,帧率有了明显的提升。
- Abstract:
- With the increasing requirements for large-scale scenes in the field of games and simulation and in order to ensure that the general computer can still run at a higher frame rate in the scene of a large-scale three-dimensional model, we propose a scene traversal algorithm based on Morton Code. It is based on the idea of full quadtree node encoding for improving the Morton Code of linear quadtree, and a linear multi-segment structure is proposed.It is also combined with bottom-up BVH to reduce traversal nodes, and finally the frustum is used to quickly crop the objects in the scene. This algorithm combines the multi-layer uniform grid in the scene, uses the CPU’s multi-core parallel computing capabilities, and quickly obtains the lower-layer input according to the upper-layer traversal results. It can quickly cut the visibility of the objects in the scene and improve the computer’s large-scale model scene rendering frame rate. Experimental results show the correctness and feasibility of the algorithm. Compared with the traditional acceleration structure, its frame rate has been significantly improved.
参考文献/References:
[1] Mohammad Mirzadeh,Arthur Guittet,CarstenBurstedde,et al.Parallel level-set methods on adaptive tree-based grids[J].Journal ofutational Physics,2016,322:345-364.
[2] Loubet G,Neyret F.Hybrid mesh-volume LoDs for all-scale pre-filtering of complex 3D assets[J].Computer Graphics Forum,2017,36(2):431-422.
[3] Argudo O,Andujar C,Chica A.Image-Based Tree Variations[J].Computer Graphics Forum,2020,39(1):174-184.
[4] Zonghui Li,Yangdong Deng,Ming Gu.Path compression kd-trees with multi-layer parallel construction a case study on ray tracing[C].Proceedings of the 21st ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games,2017,(16):1-8.
[5] MarekVinkler,Jií Bittner,VlastimilHavran,et al.Massively Parallel Hierarchical Scene Processing with Applications in Rendering[J].Computer Graphics Forum,2013,32(8):13-25.
[6] Xiaoping Liu,YeYu,JieqiongKong,et al.Scene management strategy in collaborative rendering environment[J].Journal of System Simulation,2007(1):85-88.
[7] Yu Zhang.A dynamic scene management based on Octree[J].Computer Knowledge and Technology,2015(14):54-57.
[8] 张琴,蔡勇,常伟杰.基于空间分割的局部KD树动态构建算法[J].机械工程师,2010(12):30-32.
[9] 冯立颖.碰撞检测技术研究综述[J].计算机时代,2014,(8):7-10.
[10] Henry sowizral.Scene Graphs in the New Millennium[J].Computer Graphics and Applications(S0272-1716),2000,20(1):56-57.
[11] Meister Daniel,Bittner Jiri.Parallel Locally-Ordered Clustering for Bounding Volume Hierarchy Construction[J].IEEE Transactions on Visualization & Computer Graphics,2018,24(3):1345-1353.
[12] Yi Lei,Xiaochong Tong,Yongsheng Zhang,et al.Global multi-scale grid integer coding and spatial indexing:A novel approach for big earth observation data[J].ISPRS Journal of Photogrammetry and Remote Sensing,2020,163:202-213.
[13] Zonghui Li,Yang Deng,Ming Gu.Path compression kd-trees with multi-layer parallel construction a case study on ray tracing[J].Proceedings of the 21st ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games,I3D 2017:161-168.
[14] J Hendrich,A Pospíil,D Meister,et al.Ray Classification for Accelerated BVH Traversal[J].Computer Graphics Forum,2019,38(4):49-56.
[15] Lauterbach C,Garland M,Sengupta S,et al.Fast BVH Construction on GPUs[J].Computer Graphics Forum,2009,28(2):375-384.
[16] Pantaleoni J,Luebke D.HLBVH: hierarchical LBVH construction for real-time ray tracing of dynamic geometry[C].Conference on High Performance Graphics,ACM,2010.June 25-27.
[17] 杨鑫,王天明,许端清.基于GPU的层次包围盒快速构造方法[J].浙江大学学报(工学版),2012,46(1):84-89.
备注/Memo
收稿日期:2020-11-26